Automatic Pricing and Replenishment Decision for Vegetable Products Based on XGboost Regression Modeling

Fresh supermarkets, the general freshness of vegetable products are relatively short, and the quality of sales time increases with the deterioration. For this reason, merchants must make replenishment decisions for each vegetable category on the same day without knowing the exact single product and purchase price. Reliable market demand analysis is particularly important for replenishment and pricing decisions. We select different individual vegetables for analysis, and use XGboost regression to iterate the curves of the attrition rate of individual items, and the number of sales to achieve the prediction effect. Based on this, we introduce constraints and also add decision variables at the item level, while we introduce POS particle swarm optimization in order to assess the stability as well as the optimality of the model, and finally we derive the maximum superstore benefit for each category separately.